Research Worth Reading

  • Grid Capacity Expansion under Data Centers and Electrified Manufacturing Large Loads — Develops a multi-period grid capacity expansion model that optimizes investment in generation, storage, and transmission under emerging large loads from data centers and electrified manufacturing, with hourly dispatch resolution. Systems engineers and optimization folks: this is a practical framework for planning grid infrastructure under non-traditional demand growth.

  • Tensorized Radiative Heat Transfer for a Scalable and Calibrated Building Energy Simulator — Presents sbsim, a lightweight, data-calibrated building energy simulator using tensorized finite difference methods with radiative heat transfer. Designed for advanced control strategies around demand flexibility and grid responsiveness — relevant if you’re into building controls, simulation, or grid-interactive buildings.

  • Identifiability of Low Frequency Li-ion Battery Parameters in Time Domain — Studies which low-frequency parameters in Li-ion battery equivalent circuit models can actually be identified from BMS voltage/current data during normal operation, accounting for measurement resolution and sampling constraints. Useful for anyone working on battery diagnostics, state estimation, or BMS algorithm development.

  • Energy-Optimal Thermal Management of Heat-Pump Battery Electric Vehicles — Presents a hybrid control framework for thermal management in heat-pump BEVs, coordinating compressor, coolant pumps, and cabin blower across coupled refrigerant, coolant, and air loops while enforcing comfort and temperature constraints. A solid systems control problem if you’re interested in automotive thermal systems or multi-loop control.

  • Evaluating Skill and Stability of ArchesWeather and ArchesWeatherGen under Multi-Decadal Climate Simulations — Tests whether ArchesWeather (deterministic) and ArchesWeatherGen (probabilistic flow-matching) ML weather models maintain skill and stability when run on multi-decadal climate simulations beyond their training domain. Important read for ML engineers working on weather/climate models — probes the limits of AI weather forecasting under distribution shift.

Technology & Innovation

  • Getting Electric Truck Chargers Online Faster — Breaks down the engineering and regulatory bottlenecks in deploying heavy-duty EV charging infrastructure, with practical strategies to shorten interconnection timelines for fleet operators. If you’re interested in the real-world deployment side of electrification — not just the hardware — this is worth your time.

Today’s Synthesis

Several of today’s picks converge on a single systems-level challenge: the grid is being pulled in multiple directions by electrification, and the engineering bottlenecks are as much about coordination as they are about hardware. Grid Capacity Expansion under Data Centers and Electrified Manufacturing Large Loads gives you the planning-generation-transmission optimization backbone for modeling where and when new capacity is needed as loads shift — from hyperscale compute to industrial heat pumps and fleet depots. But planning the wires is only half the problem: Getting Electric Truck Chargers Online Faster makes clear that interconnection delays and regulatory friction are the real deployment chokepoints, not charger cost. Meanwhile, Tensorized Radiative Heat Transfer for a Scalable and Calibrated Building Energy Simulator shows how calibrated, lightweight building simulators can help operators orchestrate demand flexibility at scale — turning buildings into grid-interactive assets rather than passive loads. Together, these point to a concrete opportunity for systems and ML engineers: building co-simulation and optimization tooling that couples grid expansion models with interconnection logistics and demand-side flexibility, so planners can stress-test deployment timelines against realistic regulatory delays and flexible load profiles. That’s where the software gap is right now.